product-showcase

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly allows/requests user-provided test credentials to access authenticated pages and includes browser actions like browser_type/click that would require the agent to accept and emit those secret values verbatim to the MCP, creating an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). Source: the skill’s Browser Capture Workflow (Cursor MCP) navigates to a user-provided running app URL (e.g., http://localhost:3000) and ingests screenshot/caption text from that live application via browser_navigate/browser_snapshot/browser_take_screenshot, which can include outsider-authored free text if the app content is user-generated.